In many optical systems, it is often useful to separate and/or combine light beams having different optical characteristics, e.g., polarization states, wavelengths, etc. For example, in optical data storage systems, the optical qualities of the storage medium are altered in a manner such that changes in the reflective or transmissive properties of the medium are representative of the information recorded thereon. This information is commonly transmitted to and retrieved from the optical medium using light beams produced by a laser light source.
The information recorded on the disc is retrieved from the disc by directing a laser beam onto the disc. The reflected laser beam is then directed onto the detecting surface of a photodiode or other light detector system which transforms the reflected or transmitted laser beam signal into an electrical signal. In this manner, the data stored on the disc is transferred from the disc to the laser beam and converted into an electrical signal which carries the same information recorded on the disc. This electrical signal is further processed, and ultimately results in retrieval of the computer data, audio sound, video images, etc., represented by the information recorded on the disc.
Separating and combining optical beams is useful in several ways in such optical disc systems. For example, in many optical disc memory systems, it is desirable to use a single laser source to produce both the read and write beams, thus resulting in a smaller and more compact system design. However, when using a single laser source, it then becomes necessary to be able to separate and distinguish the two beams. In addition, read only systems are often utilized in connection with optical discs which require separation of the incident and reflected read-beams. Furthermore, it is necessary to separate the two polarization states of the reflected read beam to detect the data stored on the disc.
The most common technique used to achieve these separations utilizes combinations of cube beamsplitters and right angle prisms. The cube beamsplitter has a center interface selected such that it is sensitive to the desired parameter of interest, i.e., polarization, wavelength, etc. The beamsplitter is then typically combined with one or more right angle prisms attached to selected faces of the cube used to create the desired separation/combination geometry. This type of design is often too bulky and/or too heavy for use in many minlaturized optical heads. In addition, this separation technique requires that the beam be detected by separate, or widely separated, detectors, thereby increasing the volume requirements of the optical system.
A second technique frequently used to separate light beams involves connecting a right angle prism to a large cube. The angles of the larger cube are selected such that the transmitted beam circulates around the cube and exits the cube at an angle which is offset with respect to the reflected beam. This design, too, is typically very heavy and requires too much space for incorporation into miniature optical heads, thereby making it inefficient in many applications.